Please use this identifier to cite or link to this item:

Neural Network Based Machine Condition Monitoring System

File Description SizeFormat 
uhm_ms_3857_r.pdfVersion for non-UH users. Copying/Printing is not permitted2.02 MBAdobe PDFView/Open
uhm_ms_3857_uh.pdfVersion for UH users2.02 MBAdobe PDFView/Open

Item Summary

Title: Neural Network Based Machine Condition Monitoring System
Authors: Kaneshige, Kenichi
Issue Date: May 2004
Abstract: In this paper, machine condition monitoring techniques based on multilayered feedfoward neural network (MLFFNN) where the weights in the network are updated based on nodedecoupled extended Kalman filter (NDEKF) training method are proposed. Neural network based techniques have been widely recognized as powerful approaches for condition monitoring system, and the use of NDEKF has better performances in computational complexity and memory requirement among the Kalman filtering algorithm family. The condition monitoring system detects and identifies conditions of components through the neural network based system identification of components. Sensor signals in both time and frequency domains are analyzed to show the effectiveness of the condition monitoring scheme. The performances of diagnostic tools presented in this thesis are evaluated using the cabin temperature control system that is specifically for Boeing 767 as practical application example, and the results show the effectiveness of the developed techniques.
Rights: All UHM dissertations and theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission from the copyright owner.
Appears in Collections:M.S. - Electrical Engineering

Items in ScholarSpace are protected by copyright, with all rights reserved, unless otherwise indicated.